JP2008312346A - Motor controller and electric power steering device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a motor controller which enables the accurate detection of abnormality, while effectively suppressing the occurrence of torque ripples at two-phase drive accompanying the occurrence of an energization failure phase. <P>SOLUTION: A microcomputer detects the abnormality of a control system, based on a q-axis current deviation ΔIq. Moreover, the microcomputer includes an abnormality judging part which can detect the occurrence in case that energization failure occurs in any phase of the motor. It generates a motor control signal by the execution of the phase feed back control based on a phase current command that changes in the shape of a secant curve or a cosecant curve, with a specified rotational angle as its asymptote, as to a phase other than the energization failure occurrence phase concerned, at occurrence of the energization failure phase. In this phase current feedback control, the phase current command concerned is limited into a specified range. Then, the microcomputer does not perform the judgement processing for detecting the abnormality of the above control system, in case that it is within a rotational angle range (a current limitation range) where the phase current command is limited within a specified range, at the occurrence of energization failure. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a motor control device and an electric power steering device.

  Conventionally, many motor control devices used in an electric power steering device (EPS) or the like have any phase of the motor (any of U, V, and W) due to disconnection of a power supply line or a contact failure of a drive circuit. An abnormality detection means is provided that can detect the occurrence of the abnormality when a current-carrying failure occurs. And when generation | occurrence | production of the said abnormality is detected, the structure which stops motor control rapidly and aims at fail safe is common.

However, in EPS, as the motor control stops, the steering characteristics change greatly. That is, in order for the driver to perform an appropriate steering operation, a larger steering force is required. Based on this point, conventionally, there is a motor control device that continues motor control using two phases other than the current-carrying failure phase as a current-carrying phase even when the occurrence of a current-carrying failure phase is detected as described above (for example, Patent Document 1). As a result, the application of assist force to the steering system can be continued, and an increase in the driver's burden associated with fail-safe can be avoided.
JP 200326020 A JP 2006-67731 A

  However, when the motor control is continued using the two phases other than the current supply failure occurrence phase as the current supply phase when the current supply failure phase is generated as in the above-described conventional example, for each current supply phase as shown in FIG. In the configuration in which sine wave energization is performed (the example shown in the figure is when the U phase is abnormal and the V and W phases are energized), the steering feeling is inevitably deteriorated due to the occurrence of torque ripple.

  That is, as shown in FIG. 28, when the transition of the motor current during the conventional two-phase drive is expressed in the d / q coordinate system, the q-axis current command value that is the control target value of the motor torque is constant. Regardless, the actual q-axis current value changes sinusoidally. That is, there is a problem that the motor drive is continued in a state where the original output performance cannot be obtained because the motor current corresponding to the required torque is not generated.

  Further, in many cases, detection of control system abnormality such as overcurrent due to a drive circuit failure or sensor abnormality is performed by comparing a current deviation (mainly q-axis current deviation) of the d / q coordinate system with a predetermined threshold value. (For example, refer to Patent Document 2). However, as described above, during the two-phase driving, each current value in the d / q coordinate system changes in a sine wave shape, so that a current deviation that is the basis of the abnormality determination occurs regardless of the presence or absence of the abnormality. Become. For this reason, there is a problem that abnormality cannot be detected during two-phase driving, and there is still room for improvement in this respect.

  The present invention has been made to solve the above-mentioned problems, and its purpose is to effectively suppress the occurrence of torque ripple during two-phase driving associated with the occurrence of a poorly energized phase and to provide a highly accurate abnormality. It is an object of the present invention to provide a motor control device and an electric power steering device that enable detection.

  In order to solve the above problems, the invention according to claim 1 is a motor control signal output means for outputting a motor control signal, and a drive circuit for supplying three-phase drive power to the motor based on the motor control signal. The motor control signal output means generates the motor control signal by executing a current command value calculating means for calculating a current command value and current feedback control of a d / q coordinate system based on the current command value. Motor control signal generating means for detecting the abnormality of the control system based on the current deviation of the d / q coordinate system and detecting abnormality of energization occurring in each phase of the motor, In the motor control device that executes the output of the motor control signal by using two phases other than the energization failure occurrence phase as the energization phase when occurrence of the energization failure is detected, the occurrence of the energization failure occurs. If detected, the current command value calculation means calculates a phase current command value that changes to a secant curve or a cosecant curve with a predetermined rotation angle corresponding to the energization failure occurrence phase as an asymptotic line, The motor control signal generating means generates the motor control signal by executing phase current feedback control based on the calculated phase current command value, and the motor control signal output means receives the phase current command value as a predetermined value. Limiting means for limiting within a range is provided, and the abnormality detection means is within a rotation angle range that limits the phase current command value within a predetermined range when the energization failure occurs. The gist is not to execute the determination for detecting the abnormality of the control system.

  According to the above configuration, even during the two-phase drive, the required torque is maintained except for the rotation angle range (current limit range) that limits the phase current command value within the predetermined range in the vicinity of the predetermined rotation angle corresponding to the asymptote. A corresponding motor current can be generated. As a result, even when an energization failure phase occurs, the motor control can be continued while suppressing the generation of torque ripple. Furthermore, by limiting the phase current command value, a current deviation will occur even if there is no abnormality in the control system in particular. However, if it is within the current limit range, the current deviation will be the basis. By not performing the determination process for detecting an abnormality in the control system, it is possible to effectively prevent the occurrence of erroneous detection. As a result, the abnormality of the control system can be detected with high accuracy even during the two-phase driving.

  The invention described in claim 2 includes motor control signal output means for outputting a motor control signal, and a drive circuit for supplying three-phase drive power to the motor based on the motor control signal, and the motor control signal The output means includes a current command value calculating means for calculating a current command value, a motor control signal generating means for generating the motor control signal by executing a current feedback control of a d / q coordinate system based on the current command value, an abnormality detecting means capable of detecting an abnormality in the control system by comparing a current deviation in the d / q coordinate system and a predetermined threshold and detecting an energization defect occurring in each phase of the motor; Is detected in the motor control device that executes the output of the motor control signal using the two phases other than the energization failure occurrence phase as the energization phase. In this case, the current command value calculation means calculates a phase current command value that changes into a secant curve or a cosecant curve with a predetermined rotation angle corresponding to the energization failure occurrence phase as an asymptotic line, and the motor control The signal generating means generates the motor control signal by executing phase current feedback control based on the calculated phase current command value, and the motor control signal output means keeps the phase current command value within a predetermined range. Limiting means for limiting is provided, and when the energization failure occurs, if the phase current command value is within a rotation angle range that limits the phase current command value within a predetermined range, the phase current command value is limited. The gist is to change the threshold value in order to cope with fluctuations in current deviation in the d / q coordinate system.

  According to a third aspect of the present invention, the current command value calculating means is configured so that the following formulas are determined according to the energization failure occurrence phase:

(However, θ: rotation angle, Iq *: q-axis current command value, Iu *: U-phase current command value, Iv *: V-phase current command value)
And calculating the phase current command value based on the following equation:

(However, Ix *: Phase current command value, Ix_max: Maximum value of phase current that can be energized)
And the abnormality detecting means detects an abnormality of the control system by comparing a q-axis current deviation of the d / q coordinate system with a predetermined threshold value. When the energization failure occurs and the phase current command value is within a rotation angle range that restricts the phase current command value within a predetermined range,

(Where, β: correction term, Iq *: q-axis current command value, Ix_max: maximum current that can be energized)
And calculating a correction term for changing the threshold.
According to each of the above configurations, even in the case of two-phase driving, the required torque is obtained except for the vicinity of a predetermined rotation angle corresponding to the asymptote and the rotation angle range (current limit range) that limits the phase current command value within the predetermined range. The motor current corresponding to the can be generated. As a result, even when an energization failure phase occurs, the motor control can be continued while suppressing the generation of torque ripple. Further, by limiting the phase current command value, a current deviation occurs in the d / q coordinate system even when there is no abnormality in the control system in particular. Accordingly, it is possible to avoid the occurrence of erroneous detection within the current limit range by changing the threshold value used for the abnormality determination of the control system. As a result, it is possible to detect an abnormality in the control system with high accuracy even during two-phase driving.

  The invention described in claim 4 includes motor control signal output means for outputting a motor control signal, and a drive circuit for supplying three-phase drive power to the motor based on the motor control signal, and the motor control signal The output means includes a current command value calculating means for calculating a current command value, a motor control signal generating means for generating the motor control signal by executing a current feedback control of a d / q coordinate system based on the current command value, When an abnormality in the control system is detected based on a current deviation in the d / q coordinate system, and an abnormality detecting means capable of detecting a conduction failure occurring in each phase of the motor is detected, and the occurrence of the conduction failure is detected In the motor control device that executes the output of the motor control signal using the two phases other than the conduction failure occurrence phase as the conduction phase, when the occurrence of the conduction failure is detected, The command value calculating means calculates a phase current command value that changes in a secant curve or a cosecant curve with a predetermined rotation angle corresponding to the energization failure occurrence phase as an asymptote, and the motor control signal generating means The motor control signal is generated by executing phase current feedback control based on the calculated phase current command value, and the motor control signal output unit is provided with a limiting unit that limits the phase current command value within a predetermined range. The abnormality detection means calculates a virtual current command value in the d / q coordinate system corresponding to the phase current command value in consideration of the limitation when the energization failure occurs, The gist is to detect an abnormality of the control system based on a deviation between a virtual current command value and an actual current value of the d / q coordinate system.

  According to a fifth aspect of the present invention, the current command value calculating means is configured so that the following formulas are determined according to the energization failure occurrence phase:

(However, θ: rotation angle, Iq *: q-axis current command value, Iu *: U-phase current command value, Iv *: V-phase current command value)
The phase current command value is calculated based on the abnormality detection means, the abnormality detection means,

(However, Id *: d-axis current command value, Iq *: q-axis current command value)
And calculating the d-axis current command value as the virtual current command value, and the limiting means is:

(However, Ix *: Phase current command value, Ix_max: Maximum value of phase current that can be energized)
The phase current command value is limited within a predetermined range, and the abnormality detection means is within a rotation angle range that limits the phase current command value within a predetermined range when the energization failure occurs. In the case of the following equations,

(However, Id *: d-axis current command value, Iq *: q-axis current command value, Ix_max: maximum current that can be energized)
And calculating the d-axis current command value.

  The invention according to claim 6 is characterized in that the current command value calculation means has the following formulas according to the conduction failure occurrence phase:

(However, θ: rotation angle, Iq *: q-axis current command value, Iu *: U-phase current command value, Iv *: V-phase current command value)
The phase current command value is calculated based on the abnormality detection means, the abnormality detection means,

(However, Idq *: vector of current command values in the d / q coordinate system, Iq *: q-axis current command value)
To calculate a combined vector of current command values in the d / q coordinate system as the virtual current command value, and the limiting means includes the following equation:

(However, Ix *: Phase current command value, Ix_max: Maximum value of phase current that can be energized)
The phase current command value is limited within a predetermined range, and the abnormality detection means is within a rotation angle range that limits the phase current command value within a predetermined range when the energization failure occurs. In the case of the following equations,

(However, Idq *: Composite vector of current command values in the d / q coordinate system, Ix_max: Maximum current that can be energized)
The gist is to calculate a combined vector of current command values in the d / q coordinate system.

  According to each of the above configurations, even in the case of two-phase driving, the required torque is obtained except for the vicinity of a predetermined rotation angle corresponding to the asymptote and the rotation angle range (current limit range) that limits the phase current command value within the predetermined range. The motor current corresponding to the can be generated. As a result, even when an energization failure phase occurs, the motor control can be continued while suppressing the generation of torque ripple. Further, by limiting the phase current command value, even if there is no abnormality in the control system, a current deviation occurs in the d / q coordinate system. By calculating a virtual current command value in the d / q coordinate system corresponding to, and performing a control system abnormality determination based on the deviation from the actual current value, erroneous detection within the current limit range can be prevented. It can be avoided. As a result, it is possible to detect an abnormality in the control system with high accuracy even during two-phase driving.

  The invention according to claim 7 includes motor control signal output means for outputting a motor control signal, and a drive circuit for supplying three-phase drive power to the motor based on the motor control signal, and the motor control signal The output means includes a current command value calculating means for calculating a current command value, a motor control signal generating means for generating the motor control signal by executing a current feedback control of a d / q coordinate system based on the current command value, an abnormality detecting means capable of detecting an abnormality of the control system and an energization failure occurring in each phase of the motor based on a current deviation of the d / q coordinate system, and when the occurrence of the energization failure is detected, In the motor control device that executes the output of the motor control signal using the two phases other than the energization failure occurrence phase as the energization phase, when the occurrence of the energization failure is detected, the current command value calculation means Calculates a phase current command value that changes into a secant curve or a cosecant curve with a predetermined rotation angle corresponding to the energization failure occurrence phase as an asymptotic line, and the motor control signal generating means generates the calculated phase The motor control signal is generated by executing phase current feedback control based on a current command value, and the abnormality detection means is based on a phase current deviation for at least one of the energized phases when the energization failure occurs. The gist is to execute the abnormality determination of the control system.

  According to the above configuration, even during two-phase driving, a motor current corresponding to the required torque can be generated except for a predetermined rotation angle corresponding to an asymptote. As a result, even when an energization failure phase occurs, the motor control can be continued while suppressing the generation of torque ripple. Furthermore, by limiting the phase current command value, a current deviation occurs in the d / q coordinate system even when there is no abnormality in the control system. However, as long as the phase current feedback control is performed, the phase current is controlled. Such a phenomenon does not occur in the deviation. Therefore, by performing control system abnormality determination based on the phase current deviation, it is possible to accurately detect control system abnormality while avoiding erroneous detection within the current limit range even during two-phase driving. become able to.

  The invention according to claim 8 includes motor control signal output means for outputting a motor control signal, and a drive circuit for supplying three-phase drive power to the motor based on the motor control signal, and the motor control signal The output means includes a current command value calculating means for calculating a current command value, a motor control signal generating means for generating the motor control signal by executing current feedback control based on the current command value, an abnormality in a control system, and the motor An abnormality detecting means capable of detecting an energization failure occurring in each of the phases, and when the occurrence of the energization failure is detected, two phases other than the energization failure occurrence phase are set as energization phases of the motor control signal. In the motor control apparatus that executes output, the abnormality detection means detects the abnormality of the control system based on the total value of the two-phase current values constituting the energized phase when the energization failure occurs. Performing a determination, and the gist.

  That is, even in the case of two-phase driving, if there is no abnormality in the control system according to Kirchhoff's law, the total value of the two-phase phase current values that are energized phases is “0” regardless of the energization method. Should be. Therefore, according to the above configuration, it is possible to detect an abnormality in the control system with high accuracy even during two-phase driving.

The gist of the invention described in claim 9 is an electric power steering device provided with the motor control device according to any one of claims 1 to 8.
According to the above configuration, the assist force can be continuously applied while maintaining a good steering feeling even when a poorly energized phase occurs, and high-precision abnormality detection is possible even during the two-phase drive. Will be able to.

  According to the present invention, it is possible to provide a motor control device and an electric power steering device that enable highly accurate abnormality detection while effectively suppressing the occurrence of torque ripple during two-phase driving accompanying the occurrence of a poorly energized phase. Can do.

(First embodiment)
Hereinafter, a first embodiment in which the present invention is embodied in an electric power steering apparatus (EPS) will be described with reference to the drawings.

  FIG. 1 is a schematic configuration diagram of the EPS 1 of the present embodiment. As shown in the figure, a steering shaft 3 to which a steering wheel (steering) 2 is fixed is connected to a rack 5 via a rack and pinion mechanism 4. It is converted into a reciprocating linear motion of the rack 5 by the and pinion mechanism 4. The rudder angle of the steered wheels 6 is changed by the reciprocating linear motion of the rack 5.

  Further, the EPS 1 includes an EPS actuator 10 as a steering force assisting device that applies an assist force for assisting a steering operation to the steering system, and an ECU 11 as a control unit that controls the operation of the EPS actuator 10. .

  The EPS actuator 10 of the present embodiment is a so-called rack-type EPS actuator in which a motor 12 that is a driving source thereof is arranged coaxially with the rack 5, and an assist torque generated by the motor 12 is a ball screw mechanism (not shown). Is transmitted to the rack 5 via. Note that the motor 12 of the present embodiment is a brushless motor, and rotates by receiving supply of three-phase (U, V, W) driving power from the ECU 11. And ECU11 as a motor control apparatus controls the assist force given to a steering system by controlling the assist torque which this motor 12 generate | occur | produces (power assist control).

  In the present embodiment, a torque sensor 14 and a vehicle speed sensor 15 are connected to the ECU 11. Then, the ECU 11 executes the operation of the EPS actuator 10, that is, power assist control, based on the steering torque τ and the vehicle speed V detected by the torque sensor 14 and the vehicle speed sensor 15, respectively.

Next, the electrical configuration of the EPS of this embodiment will be described.
FIG. 2 is a control block diagram of the EPS of this embodiment. As shown in the figure, the ECU 11 includes a microcomputer 17 as motor control signal output means for outputting a motor control signal, and a drive circuit 18 for supplying three-phase drive power to the motor 12 based on the motor control signal. ing.

  The drive circuit 18 of this embodiment is a known PWM inverter in which three arms corresponding to each phase are connected in parallel with a pair of switching elements connected in series as a basic unit (arm). The output of the motor control signal defines the on-duty ratio of each switching element constituting the drive circuit 18. Then, a motor control signal is applied to the gate terminal of each switching element, and each switching element is turned on / off in response to the motor control signal, so that the DC voltage of the in-vehicle power supply (not shown) becomes three-phase (U, V, W) is converted into drive power and supplied to the motor 12.

  In the present embodiment, the ECU 11 includes current sensors 21u, 21v, 21w for detecting the phase current values Iu, Iv, Iw energized to the motor 12, and a rotation for detecting the rotation angle θ of the motor 12. An angle sensor 22 is connected. Then, the microcomputer 17 sends a motor to the drive circuit 18 based on the phase current values Iu, Iv, Iw and the rotation angle θ of the motor 12 detected based on the output signals of these sensors, and the steering torque τ and the vehicle speed V. Output a control signal.

  More specifically, in this embodiment, the microcomputer 17 normally converts the detected phase current values Iu, Iv, and Iw into a d-axis current value Id and a q-axis current value Iq in the d / q coordinate system, A motor control signal is generated by executing current feedback control in the d / q coordinate system.

  Specifically, the microcomputer 17 includes a current command value calculation unit 23 as a current command value calculation unit that calculates a current command value as an assist force to be applied to the steering system, that is, a motor torque control target value, and a current command value calculation. And a motor control signal generation unit 24 as a motor control signal generation unit that generates a motor control signal based on the current command value calculated by the unit 23.

  The current command value calculation unit 23 calculates the d-axis current command value Id * and the q-axis current command value Iq * based on the steering torque τ and the vehicle speed V detected by the torque sensor 14 and the vehicle speed sensor 15 in normal times. And output to the motor control signal generator 24. On the other hand, the motor control signal generator 24 detects the d-axis current command value Id * and the q-axis current command value Iq * calculated by the current command value calculator 23 by the current sensors 21u, 21v, and 21w. Each phase current value Iu, Iv, Iw and the rotation angle θ detected by the rotation angle sensor 22 are input. Then, the motor control signal generation unit 24 executes current feedback control in the d / q coordinate system based on the phase current values Iu, Iv, Iw and the rotation angle θ (electrical angle), thereby performing the motor control signal. Is generated.

  More specifically, the motor control signal generator 24 of the present embodiment calculates a first control unit 24a that calculates three-phase phase voltage command values Vu *, Vv *, and Vw * by current feedback control in the d / q coordinate system. In normal times, the motor control signal is generated based on the phase voltage command values Vu *, Vv *, Vw * calculated by the first control unit 24a.

  Specifically, in the first control unit 24a, the phase current values Iu, Iv, and Iw are input to the three-phase / two-phase conversion unit 25 together with the rotation angle θ, and the three-phase / two-phase conversion unit 25 d / Q coordinate system to d-axis current value Id and q-axis current value Iq. Further, the q-axis current command value Iq * output from the current command value calculation unit 23 is input to the subtractor 26q together with the q-axis current value Iq, and the d-axis current command value Id * is subtracted together with the d-axis current value Id. Is input to the device 26d. In the present embodiment, “0” is input as the d-axis current command value Id * to the subtractor 26d in a normal state (Id * = 0). The d-axis current deviation ΔId and q-axis current deviation ΔIq calculated in the subtractors 26d and 26q are input to the corresponding F / B control units 27d and 27q, respectively. In these F / B control units 27d and 27q, the d-axis current command value Id * and the q-axis current command value Iq * output from the current command value calculation unit 23 are the actual d-axis current values Id and q Feedback control for following the shaft current value Iq is performed.

  That is, the F / B control units 27d and 27q multiply the inputted d-axis current deviation ΔId and q-axis current deviation ΔIq by a predetermined F / B gain (PI gain), thereby obtaining a d-axis voltage command value Vd * and The q-axis voltage command value Vq * is calculated. The d-axis voltage command value Vd * and the q-axis voltage command value Vq * calculated by the F / B control units 27d and 27q are input to the two-phase / three-phase conversion unit 28 together with the rotation angle θ. The two-phase / three-phase conversion unit 28 converts the three-phase voltage command values Vu *, Vv *, and Vw *.

  In the present embodiment, the phase voltage command values Vu *, Vv *, Vw * calculated in the first control unit 24a as described above are input to the PWM conversion unit 30 via the switching control unit 29, and the PWM The converter 30 generates duty command values αu, αv, αw corresponding to the phase voltage command values Vu *, Vv *, Vw *. The motor control signal generator 24 generates a motor control signal having an on-duty ratio indicated by each of the duty command values αu, αv, αw, and the microcomputer 17 configures the drive circuit 18 with the motor control signal. By outputting to each switching element (the gate terminal thereof), the operation of the drive circuit 18, that is, the supply of drive power to the motor 12 is controlled.

[Control mode when an abnormality occurs]
As shown in FIG. 2, in the ECU 11 of the present embodiment, the microcomputer 17 is provided with an abnormality determination unit 31 for identifying the abnormality mode when any abnormality occurs in the EPS 1. Then, the ECU 11 (microcomputer 17) changes the control mode of the motor 12 according to the abnormality mode specified (determined) by the abnormality determination unit 31.

  More specifically, an abnormality signal S_tr for detecting an abnormality in the mechanical system of the EPS actuator 10 is input to the abnormality determination unit 31, and the abnormality determination unit 31 receives the input abnormality signal. Based on S_tr, an abnormality in the mechanical system in EPS 1 is detected. Further, the abnormality determination unit 31 includes a q-axis current command value Iq * and a q-axis current value Iq, each phase current value Iu, Iv, Iw of the motor 12, a rotational angular velocity ω, and a duty command value αu, αv for each phase. , Αw, etc. are input. And the abnormality determination part 31 detects generation | occurrence | production of abnormality in a control system based on each of these state quantities.

  Specifically, the abnormality determination unit 31 of the present embodiment monitors the q-axis current deviation ΔIq in order to detect the occurrence of an abnormality related to the entire control system, such as a failure of the torque sensor 14 or a failure of the drive circuit 18. That is, the q-axis current deviation ΔIq is compared with a predetermined threshold value, and if the q-axis current deviation ΔIq is equal to or greater than the threshold value (continuous for a predetermined time), it is determined that an abnormality has occurred in the control system. To do.

  Further, the abnormality determination unit 31 disconnects or drives a power line (including a motor coil) based on the phase current values Iu, Iv, Iw, the rotational angular velocity ω, and the duty command values αu, αv, αw of each phase. The occurrence of an energization failure phase caused by a contact failure of the circuit 18 is detected. The detection of the occurrence of a poorly energized phase is performed by detecting the phase current value Ix of the X phase (X = U, V, W) below the predetermined value Ith (| Ix | ≦ Ith) and the rotational angular velocity ω within the target range for disconnection determination ( When | ω | ≦ ω0), the state in which the duty command value αx corresponding to the phase is not in the predetermined range (αLo ≦ αx ≦ αHi) corresponding to the predetermined value Ith and the threshold value ω0 defining the determination target range continues. Depending on whether or not.

  In this case, the predetermined value Ith serving as the threshold value of the phase current value Ix is set to a value near “0”, and the threshold value ω0 of the rotational angular velocity ω is a value corresponding to the base speed (maximum rotational speed) of the motor. Set to The threshold values (αLo, αHi) relating to the duty command value αx are set to a value smaller than a lower limit value that can be taken by the duty command value αx and a value larger than the upper limit value in normal control.

  That is, as shown in the flowchart of FIG. 3, the abnormality determination unit 31 determines whether or not the detected phase current value Ix (absolute value thereof) is equal to or less than the predetermined value Ith (step 101), and determines the predetermined value Ith. If it is below (| Ix | ≦ Ith, Step 101: YES), it is subsequently determined whether or not the rotational angular velocity ω (the absolute value thereof) is not more than a predetermined threshold value ω0 (Step 102). When the rotational angular velocity ω is equal to or less than the predetermined threshold ω0 (| ω | ≦ ω0, step 102), it is determined whether or not the duty command value αx is within the predetermined range (αLo ≦ αx ≦ αHi). If it is determined (step 103) and it is not within the predetermined range (step 103: NO), it is determined that an energization failure has occurred in the X phase (step 104).

  When the phase current value Ix is larger than the predetermined value Ith (| Ix |> Ith, step 101: NO), when the rotational angular velocity ω is larger than the threshold ω0 (| ω |> ω0, step 102: NO), Alternatively, if the duty command value αx is within the predetermined range (αLo ≦ αx ≦ αHi, step 103: YES), it is determined that no energization failure has occurred in the X phase (normal X phase, step 105).

  That is, when an energization failure (disconnection) occurs in the X phase (any one of the U, V, and W phases), the phase current value Ix of the phase is “0”. Here, in the case where the X-phase phase current value Ix is “0” or “a value close to 0”, there may be the following two cases in addition to the occurrence of such disconnection.

− When the rotational angular speed of the motor reaches the base speed (maximum rotational speed) − When the current command itself is substantially “0” Based on this point, in this embodiment, first, the phase current of the X phase that is the determination target By comparing the value Ix with a predetermined value Ith, it is determined whether or not the phase current value Ix is “0”. Then, it is determined whether or not the above two cases where the phase current value Ix is “0” or “a value close to 0” other than when the wire is disconnected. It is determined that a disconnection has occurred.

  That is, when the extreme duty command value αx is output even though the rotational angular velocity ω (base velocity) is not such that the phase current value Ix is equal to or less than the predetermined value Ith in the vicinity of “0”, the X It can be determined that a current conduction failure has occurred in the phase. And in this embodiment, it has the structure which specifies the phase in which the conduction failure generate | occur | produced by performing the said determination sequentially about each phase of U, V, and W. FIG.

  Although omitted in the flowchart of FIG. 3 for convenience of explanation, the above determination is made on the assumption that the power supply voltage is equal to or higher than a specified voltage necessary for driving the motor 12. Then, the final abnormality detection determination is made based on whether or not the state in which it is determined that the energization failure has occurred in the predetermined step 104 has continued for a predetermined time or more.

  In the present embodiment, the ECU 11 (microcomputer 17) switches the control mode of the motor 12 based on the result of the abnormality determination in the abnormality determination unit 31. Specifically, the abnormality determination unit 31 outputs the result of abnormality determination including the above-described failure of energization as an abnormality detection signal S_tm, and the current command value calculation unit 23 and the motor control signal generation unit 24 receive the input. The calculation of the current command value according to the abnormality detection signal S_tm to be performed and the generation of the motor control signal are executed. As a result, the control mode of the motor 12 in the microcomputer 17 can be switched.

  More specifically, the ECU 11 of the present embodiment is a “normal control mode” that is a normal control mode, and a “assist stop mode” that is a control mode when an abnormality that should stop driving the motor 12 occurs. , And “two-phase drive mode” which is a control mode in the case where energization failure occurs in any of the phases of the motor 12, and the above three broad control modes. When the abnormality detection signal S_tm output from the abnormality determination unit 31 corresponds to the “normal control mode”, the current command value calculation unit 23 and the motor control signal generation unit 24 perform the normal operation as described above. The d-axis current command value Id * and q-axis current command value Iq * are calculated and the motor control signal is generated.

  On the other hand, when the abnormality detection signal S_tm output from the abnormality determination unit 31 is in the “assist stop mode”, the current command value calculation unit 23 and the motor control signal generation unit 24 perform d d to stop driving the motor 12. Calculation of the shaft current command value Id * and q-axis current command value Iq * and generation of a motor control signal are executed. The “assist stop mode” is selected not only when an abnormality occurs in the mechanical system or the torque sensor 14, but when an abnormality occurs in the power supply system, an overcurrent may occur. Can be mentioned. Further, in the “assist stop mode”, there is a case where the output of the motor 12 is gradually reduced, that is, the assist force is gradually reduced and then stopped, in addition to the case where the drive of the motor 12 is immediately stopped. The motor control signal generator 24 gradually reduces the value (absolute value) of the q-axis current command value Iq * that is output. Then, after the motor 12 is stopped, the microcomputer 17 is configured to open each switching element constituting the drive circuit 18 and open a power relay (not shown).

  Further, the abnormality detection signal S_tm corresponding to the “two-phase drive mode” includes information for specifying the energization failure occurrence phase. When the abnormality detection signal S_tm output from the abnormality determination unit 31 corresponds to this “two-phase drive mode”, the motor control signal generation unit 24 sets two phases other than the current-carrying failure occurrence phase as current-carrying phases. The motor control signal to be generated is executed.

  Here, in this embodiment, in this “two-phase drive mode”, the current command value calculation unit 23 controls the required torque, that is, the motor torque, except for a predetermined rotation angle corresponding to the phase in which the conduction failure has occurred. A phase current command value that generates a motor current (q-axis current value Iq) corresponding to the target value (q-axis current command value Iq *) is calculated. Then, the motor control signal generator 24 generates a motor control signal by executing phase current feedback control based on the phase current command value.

  More specifically, the current command value calculation unit 23 of the present embodiment receives the detected abnormality occurrence phase (when the abnormality detection signal S_tm corresponding to the “two-phase drive mode” is input from the abnormality determination unit 31. Based on the following (1) to (3) formulas, a phase current command value of one phase of the remaining two phases is calculated according to the energization failure occurrence phase.

  That is, by calculation based on these equations, a predetermined curve (the reciprocal of cos θ (secant: secθ)) or a cosecant curve (with a predetermined rotation angle θA, θB corresponding to the phase in which the conduction failure has occurred is used as an asymptote. A phase current command value that changes in a reciprocal of sin θ (cosecant: cosec θ) is calculated. In this paper, for convenience of explanation, of the two rotation angles corresponding to the asymptotes described above, in the electric angle range of 0 ° to 360 °, the smaller value is the rotation angle θA, and the larger rotation angle is the rotation angle. This will be described as θB.

  Specifically, as shown in FIG. 4, when the two phases V and W are energized due to abnormality of the U phase, the predetermined rotation angles θA and θB are “90 °” and “270 °”, respectively. . Further, as shown in FIG. 5, the predetermined rotation angles θA and θB when the V phase is a current-carrying failure generation phase are “30 °” and “210 °”, respectively. As shown in FIG. 6, the predetermined rotation angles θA and θB in the case where the W phase is an energization failure occurrence phase are “150 °” and “330 °”, respectively.

  Actually, since there is an upper limit to the current (absolute value) that can be applied to the motor coils 12u, 12v, 12w of each phase, in this embodiment, the current is calculated by the above equations (1) to (3). The phase current command value is subjected to guard processing for limiting the value within a predetermined range (described later). Therefore, as shown in FIGS. 4 to 6, the phase current command value can be energized within the range in which the guard process is performed (current limiting range: θ1 <θ <θ2, θ3 <θ <θ4). It becomes constant at the upper limit or lower limit.

  Then, the phase current feedback control is executed based on the phase current command value that changes into a secant curve or a cosecant curve with the predetermined rotation angles θA and θB as asymptotic lines, and a motor control signal is generated. In addition, the motor current (q-axis current value Iq) corresponding to the required torque (q-axis current command value Iq *) can be generated except for the predetermined rotation angles θA and θB.

  That is, as shown in FIGS. 7 to 9, in the configuration in which the guard process is performed on the phase current command value as in the present embodiment, the current limiting range (θ1 near the predetermined rotation angles θA and θB corresponding to the asymptotes). Except for <θ <θ2, θ3 <θ <θ4), a motor current (q-axis current value Iq) corresponding to the required torque (q-axis current command value Iq *) can be generated. And in this embodiment, it becomes the structure which continues provision | providing of assist force by this, suppressing generation | occurrence | production of a torque ripple and maintaining a favorable steering feeling also at the time of the electricity supply failure phase generation | occurrence | production.

  More specifically, as shown in FIG. 2, the motor control signal generation unit 24 of the present embodiment has a phase current command value (Ix *: X = U, V, W) output from the current command value calculation unit 23. The second control unit 24b that calculates the three-phase phase voltage command values Vu **, Vv **, and Vw ** by executing the phase current feedback control based on (i). When the abnormality detection signal S_tm corresponding to the “two-phase drive mode” is input from the abnormality determination unit 31, the motor control signal generation unit 24 calculates each phase voltage calculated in the second control unit 24 b. A motor control signal is generated based on the command values Vu **, Vv **, and Vw **.

  That is, the second control unit 24b of the present embodiment performs the phase current feedback control on one phase (control phase) of the two phases other than the energization failure occurrence phase, and is obtained by executing the phase current feedback control. Based on the phase voltage command value for the control phase, each phase voltage command value Vu **, Vv **, Vw ** is calculated.

  Specifically, the second control unit 24b includes a control phase selection unit 32 to which the detected phase current values Iu, Iv, Iw and the abnormality detection signal S_tm are input. 32, based on the abnormality detection signal S_tm, selects a phase for executing phase current feedback control from one of the two phases other than the energization failure occurrence phase, that is, a control phase.

  On the other hand, in the present embodiment, the phase current command value Ix * output from the current command value calculation unit 23 is input to the guard processing unit 33 as a limiting unit, and the guard processing unit 33 A guard process is performed to limit the input phase current command value Ix * within a predetermined range.

  Specifically, the guard processing unit 33 limits the output phase current command value Ix ** (the absolute value thereof) after the guard processing to the range expressed by the following equation (4).

  In the above equation (4), “Ix_max” is the maximum value of the current value that can be supplied to the X phase (U, V, W phase), and this maximum value is the switching element that constitutes the drive circuit 18. It is defined by the rated current. The equation (4) is a relational equation regarding the condition that the phase current command value Ix ** should satisfy when such a maximum value exists.

  The phase current command value Ix ** after the guard processing is performed in the guard processing unit 33 is input to the subtractor 34 together with the phase current value Ix selected as the control phase in the control phase selection unit 32. The subtractor 34 calculates the phase current deviation ΔIx by subtracting the phase current value Ix from the phase current command value Ix *, and outputs the calculated phase current deviation ΔIx to the F / B control unit 35. Then, the F / B control unit 35 calculates the phase voltage command value Vx * for the control phase by multiplying the input phase current deviation ΔIx by a predetermined F / B gain (PI gain).

  The phase voltage command value Vx * calculated by the F / B control unit 35 is input to the phase voltage command value calculation unit 36. Then, the phase voltage command value calculation unit 36 calculates each phase voltage command value Vu **, Vv **, Vw ** based on the phase voltage command value Vx * for the control phase.

  That is, of course, the energization failure occurrence phase cannot be energized, and the phase of the two energization phases is shifted by π / 2 (180 °) during two-phase driving. Accordingly, the phase voltage command value of the phase where the power failure has occurred is “0”, and the phase voltage command value of the other current phase can be calculated by inverting the sign of the phase voltage command value Vx * related to the control phase. Then, the second control unit 24 b outputs the phase voltage command values Vu **, Vv **, and Vw ** calculated in this way to the switching control unit 29.

  An abnormality detection signal S_tm is input to the switching control unit 29. When the abnormality detection signal S_tm indicates the “two-phase drive mode”, the switching control unit 29 performs the first control. Instead of the phase voltage command values Vu *, Vv * and Vw * output from the unit 24a, the phase voltage command values Vu **, Vv ** and Vw ** output from the second control unit 24b are converted into PWM converters. Output to 30. A motor control signal having an on-duty ratio corresponding to each phase voltage command value Vu **, Vv **, Vw ** is generated and output to the drive circuit 18.

Next, the abnormality determination and control mode switching by the microcomputer and the motor control signal generation processing procedure during two-phase driving will be described.
As shown in the flowchart of FIG. 10, the microcomputer 17 first determines whether or not any abnormality has occurred (step 201). If it is determined that an abnormality has occurred (step 201: YES), then It is determined whether the abnormality is a control system abnormality (step 202). Next, when it is determined in step 202 that a control system abnormality has occurred (step 202: YES), it is determined whether or not the current control mode is the two-phase drive mode (step 203), and two-phase drive is performed. If it is not the mode (step 203: NO), it is determined whether or not the abnormality of the control system is the occurrence of a poorly energized phase (step 204). If it is determined that a current-carrying failure phase has occurred (step 204: YES), a motor control signal is output that uses the remaining two phases other than the current-carrying failure phase as a current-carrying phase (two-phase drive mode, step). 205).

  As described above, the output of the motor control signal in the two-phase drive mode is a phase current command that changes into a secant curve or a remainder curve with predetermined rotation angles θA and θB corresponding to the energization failure occurrence phase as asymptotic lines. This is performed by calculating a value and executing phase current feedback control based on the phase current command value.

  That is, as shown in the flowchart of FIG. 11, the microcomputer 17 first determines whether or not the current-carrying failure occurrence phase is the U phase (step 301), and if it is the U phase (step 301: YES). In step S302, the phase current command value Iv * for the V phase is calculated based on the equation (1). Next, the microcomputer 17 executes a guard process calculation for the phase current command value Iv *, and limits the phase current command value Iv ** after the guard process within a predetermined range (step 303). Then, by executing phase current feedback control based on the phase current command value Iv ** after the guard processing, a phase voltage command value Vv * for the V phase is calculated (step 304), and based on the phase voltage command value Vv *. Thus, phase voltage command values Vu **, Vv **, and Vw ** for each phase are calculated (Vu ** = 0, Vv ** = Vv *, Vw ** =-Vv *, step 305).

  On the other hand, if it is determined in step 301 that the energization failure occurrence phase is not the U phase (step 301: NO), the microcomputer 17 determines whether the energization failure occurrence phase is the V phase (step 306). When the failure occurrence phase is the V phase (step 306: YES), the phase current command value Iu * for the U phase is calculated based on the above equation (2) (step 307). Next, the microcomputer 17 executes a guard processing calculation for the phase current command value Iu *, and limits the phase current command value Iu ** after the guard processing within a predetermined range (step 308). Then, phase current feedback control based on the phase current command value Iu ** after the guard processing is executed (step 309), and each phase voltage command value Vu * calculated by the execution of the phase current feedback control is The phase voltage command values Vu **, Vv **, and Vw ** of the phases are calculated (Vu ** = Vu *, Vv ** = 0, Vw ** = − Vu *, step 310).

  If it is determined in step 306 that the energization failure occurrence phase is not the V phase (step 306: NO), the microcomputer 17 determines the phase current command value Iv * for the V phase based on the above equation (3). Is calculated (step 311), and then the guard process calculation is executed to limit the phase current command value Iv ** after the guard process within a predetermined range (step 312). Then, phase current feedback control based on the phase current command value Iv ** after the guard processing is executed (step 313), and each phase voltage command value Vv * calculated by executing the phase current feedback control is The phase voltage command values Vu **, Vv **, and Vw ** of the phases are calculated (Vu ** = − Vv *, Vv ** = Vv *, Vw ** = 0, step 314).

  The microcomputer 17 generates a motor control signal based on the phase voltage command values Vu **, Vv **, and Vw ** calculated in step 305, step 310, or step 314, and outputs the motor control signal to the drive circuit 18. (Step 315).

  If it is determined in step 201 that there is no abnormality (step 201: NO), as described above, the microcomputer 17 executes the current feedback control in the d / q coordinate system to execute the motor control signal. The output is executed (normal control mode, step 206). Further, when it is determined in step 202 that an abnormality other than the control system has occurred (step 202: NO), when it is determined in step 203 that the two-phase drive mode has already been established (step 203: YES), or the above If it is determined in step 203 that an abnormality other than the occurrence of a poorly energized phase has occurred (step 203: NO), the microcomputer 17 shifts to the assist stop mode (step 207). And the output of the motor control signal for stopping the drive of the motor 12, opening of the power relay, etc. are executed.

[Control system error detection]
Next, the control system abnormality detection mode in this embodiment will be described.
As described above, in the present embodiment, the output of the motor control signal in the two-phase drive mode changes to a secant curve or a remainder curve with the predetermined rotation angles θA and θB corresponding to the energization failure occurrence phase as asymptotic lines. This is performed by calculating a phase current command value to be executed and executing phase current feedback control based on the phase current command value (see FIGS. 4 to 6). Thus, theoretically, the motor current (q-axis current value Iq) corresponding to the required torque (q-axis current command value Iq *) can be generated except for the predetermined rotation angles θA and θB (FIG. 7 to FIG. 7). As a result, the generation of torque ripple can be suppressed, and the application of assist force can be continued while maintaining a good steering feeling. Furthermore, the abnormality determination unit 31 of the present embodiment detects the occurrence of an abnormality in the control system based on a comparison between the q-axis current deviation ΔIq and a predetermined threshold value I1. Therefore, the motor current (q-axis current value Iq) corresponding to the required torque (q-axis current command value Iq *) is generated as described above, and theoretically in two-phase driving (two-phase driving mode). As a result, it becomes possible to detect abnormalities in the control system.

  However, in reality, there is an upper limit for the current (absolute value) that can be applied to the motor coils 12u, 12v, 12w of each phase, and the value is set for each phase current command value calculated during two-phase driving. A guard process is performed to limit the range within a predetermined range (see FIGS. 4 to 6). Therefore, as shown in FIGS. 7 to 9, within the current limiting range in the vicinity of the predetermined rotation angles θA and θB (θ1 <θ <θ2, θ3 <θ <θ4), the q axis As a result, a current deviation ΔIq is generated, which may cause a false detection.

  In view of this point, in the present embodiment, the abnormality determination unit 31 performs the determination for detecting the abnormality of the control system as described above, specifically, at the time of occurrence of a poorly energized phase, that is, at the time of two-phase driving. The determination process based on the q-axis current deviation ΔIq is not executed.

  In other words, the current limit range is a range (Ix * ≧ Ix_max, Ix * ≦ −Ix_max) in which the absolute value of the phase current command value Ix * is equal to or greater than the maximum value Ix_max of the energizable current. Therefore, an upper limit value (+ Ix_max) and a lower limit value (−Ix_max) corresponding to the maximum value Ix_max are substituted into the expressions (1) to (3), and the current is obtained by solving these expressions with respect to the rotation angle θ. It is possible to specify a limited range.

  Specifically, when the U phase is an energization failure occurrence phase (when the U phase energization failure occurs), the upper limit value and lower limit value of the energizable current are substituted into the above equation (1) (5) (6) And (7) By solving the equation (8), one of the predetermined rotation angles θA and θB corresponding to the asymptote, and the rotation angles θ1 and θ2 defining the current limiting range in the vicinity of the rotation angle θA can be obtained. it can.

  Here, the solutions of the above equations (5) to (8), that is, the rotation angles θ1 and θ2 are represented by an inverse function (arc cosine) of cos θ. Therefore, from cos (−θ) = cosθ and cos (θ + 2π) = cosθ, the current limiting range in the vicinity of the rotation angle θB, that is, the predetermined rotation angle on the other side whose phase is shifted by 2π from the rotation angle θA is defined. The rotation angles θ3 and θ4 can be obtained by the following equations (9) and (10).

  Similarly, when the V phase is an energization failure occurrence phase (at the time of V phase energization failure), the upper limit value and the lower limit value of the energizable current are substituted into the above equation (2) (11) (12) And, by solving the equations (13) and (14), one of the predetermined rotation angles θA and θB corresponding to the asymptote, and the rotation angles θ1 and θ2 defining the current limiting range in the vicinity of the rotation angle θA can be obtained. it can.

  Here, the solutions of the above equations (11) to (14), that is, the rotation angles θ1 and θ2 are expressed by an inverse function (arc sine) of sin θ. Therefore, from sin (π−θ) = sinθ, the rotation angles θ1 and θ2 that define the current limiting range in the vicinity of the rotation angle θA, and the predetermined rotation angle on the other side whose phase is shifted by 2π from the rotation angle θA, that is, The following equations (15) and (16) are established between the rotation angles θ3 and θ4 that define the current limiting range in the vicinity of the rotation angle θB. The remaining rotation angles θ3 and θ4 can be obtained by solving these equations (15) and (16).

  Further, when the W phase is an energization failure occurrence phase (at the time of W phase energization failure), the upper limit value and the lower limit value of the energizable current are substituted into the above equation (3) (17) (18) and (19) By solving the equation (20), one of the predetermined rotation angles θA and θB corresponding to the asymptote, and the rotation angles θ3 and θ4 that define the current limiting range in the vicinity of the rotation angle θB can be obtained. .

  Here, the solutions of the above equations (17) to (20), that is, the rotation angles θ3 and θ4 are expressed by an inverse function (arc sine) of sin θ. Therefore, from sin (π−θ) = sinθ, the rotation angles θ3 and θ4 that define the current limiting range in the vicinity of the rotation angle θB, and the predetermined rotation angle on the other side whose phase is shifted by 2π from the rotation angle θB, that is, The following equations (21) and (22) are established between the rotation angles θ1 and θ2 that define the current limiting range in the vicinity of the rotation angle θA.

The remaining rotation angles θ1 and θ2 can be obtained by solving these equations (21) and (22).
As described above, the abnormality determination unit 31 according to the present embodiment calculates the rotation angles θ1, θ2, θ3, and θ4 that define the current limiting ranges in the vicinity of the predetermined rotation angles θA and θB corresponding to the two asymptotes. When the detected rotation angle θ of the motor 12 is within any current limit range (θ1 <θ <θ2 or θ3 <θ <θ4), a determination is made to detect an abnormality in the control system. do not do. As a result, it is possible to prevent occurrence of erroneous detection and to detect an abnormality of the control system with high accuracy even during two-phase driving.

  More specifically, as shown in the flowchart of FIG. 12, the abnormality determination unit 31 uses the q-axis current command value Iq *, the q-axis current value Iq, and the rotation angle θ as state quantities used for abnormality detection of the control system. Upon acquisition (step 401), first, the q-axis current deviation ΔIq is calculated (ΔIq = Iq * −Iq, step 402). Next, the abnormality determination unit 31 determines whether or not the measurement flag is “ON” (step 403), and when the measurement flag is not “ON” (measurement flag = “OFF”, step 403: NO). Initializes a counter for measuring elapsed time (t = 0, step 404). When it is determined in step 403 that the measurement flag is already “ON” (step 403: YES), the abnormality determination unit 31 does not execute the counter initialization process in step 404.

  Next, the abnormality determination unit 31 determines whether or not the current control mode is the two-phase drive mode (during two-phase drive) (step 405), and when the two-phase drive is not being performed (step 405: NO). That is, when it is determined that the normal control is being performed (during the normal control mode stay), it is determined whether or not the q-axis current deviation ΔIq (absolute value thereof) is equal to or greater than a predetermined threshold value I1 (step 406). If the q-axis current deviation ΔIq (absolute value) is equal to or greater than a predetermined threshold value I1 (| ΔIq | ≧ I1, Step 406: YES), the measurement flag is set to “ON” (Step 407), and the counter Is incremented (t = t + 1, step 408). In step 406, when the q-axis current deviation ΔIq (absolute value thereof) is less than the predetermined threshold value I1 (| ΔIq | <I1, step 406: NO), the abnormality determination unit 31 sets the measurement flag. “OFF” is set (step 409).

  Next, the abnormality determination unit 31 determines whether or not the counter value t is equal to or greater than a predetermined threshold value t1 (step 410). If it is determined that the counter value t is equal to or greater than the predetermined threshold value t1 (step 410: YES), it is determined that an abnormality has occurred in the control system (step 411).

  That is, the abnormality determination unit 31 of the present embodiment sets the measurement flag to “ON” when the value of the q-axis current deviation ΔIq indicates the occurrence of abnormality in the control system (step 406: YES), and counts the duration time. To start. If the state indicating the abnormality continues for a predetermined time or longer (step 410: YES), it is determined that an abnormality has occurred in the control system.

  On the other hand, when it is determined in step 405 that the two-phase driving is already being performed (step 405: YES), the abnormality determination unit 31 corresponds to the two asymptotes according to the expressions (5) to (22). The rotation angles θ1, θ2, θ3, and θ4 that define the current limit ranges in the vicinity of the predetermined rotation angles θA and θB are calculated (current limit range calculation, step 412). Then, it is determined whether or not the detected rotation angle θ is within any current limiting range defined by each of these rotation angles θ1, θ2 or θ3, θ4 (step 413), and any current limiting range is determined. The determination process of step 406 is executed only when the above does not apply (0 <θ ≦ θ1, θ2 ≦ θ ≦ θ3, or θ4 ≦ θ ≦ 2π, step 413: NO). That is, the abnormality determination unit 31 determines that the rotation angle θ is in any current limiting range in the above step 412 (θ1 <θ <θ2 or θ3 <θ <θ4, step 413: NO). Does not execute the determination process related to the abnormality detection of the control system (step 414). Then, the processing from step 406 to step 409 is executed without executing the processing from step 410.

As described above, according to the present embodiment, the following operations and effects can be obtained.
The microcomputer 17 detects an abnormality in the control system based on a comparison between the q-axis current deviation ΔIq and a predetermined threshold value I1. Further, the microcomputer 17 includes an abnormality determination unit 31 that can detect the occurrence of an energization failure in any phase of the motor 12, and when an energization failure phase occurs, a phase other than the energization failure occurrence phase is detected. The motor control signal is generated by executing the phase current feedback control based on the phase current command value that changes into a secant curve or a cosecant curve with a predetermined rotation angle as an asymptotic line. During this phase current feedback control, the phase current command value is limited within a predetermined range. Then, the microcomputer 17 determines that the abnormality in the control system is to be detected if the phase current command value is within a rotation angle range (current limit range) that is limited within a predetermined range when the energization failure occurs. Do not process.

  According to the above configuration, the required torque (q-axis current command value Iq *) is excluded except for the current limiting ranges in the vicinity of the predetermined rotation angles θA and θB corresponding to the asymptote. ) Can be generated (q-axis current value Iq). As a result, even when an energization failure phase occurs, it is possible to continue the application of the assist force while suppressing the generation of torque ripple and maintaining a good steering feeling. Further, in the current limiting range where the q-axis current deviation ΔIq occurs regardless of whether or not there is a control system abnormality, the determination process for detecting the control system abnormality based on the q-axis current deviation ΔIq is not performed. Therefore, it is possible to effectively prevent the occurrence of erroneous detection. As a result, the abnormality of the control system can be detected with high accuracy even during the two-phase driving.

(Second Embodiment)
Hereinafter, a second embodiment in which the present invention is embodied in an electric power steering device (EPS) will be described with reference to the drawings. Note that the main difference between the present embodiment and the first embodiment is only the manner of detecting abnormality in the control system. For this reason, for convenience of explanation, the same parts as those in the first embodiment are denoted by the same reference numerals, and the explanation thereof is omitted.

  As described above, by performing the guard process that limits the phase current command value within a predetermined range in the vicinity of the predetermined rotation angles θA and θB corresponding to the asymptote (see FIGS. 4 to 6), the predetermined Within the current limiting range in the vicinity of the rotation angles θA, θB (θ1 <θ <θ2, θ3 <θ <θ4), a q-axis current deviation ΔIq occurs regardless of whether there is an abnormality (see FIGS. 7 to 9). ). For this reason, in the first embodiment, in the current limit range, the detection process based on the q-axis current deviation ΔIq is not subjected to the determination process to detect the occurrence of false detection. It was decided to.

  On the other hand, in the present embodiment, the abnormality determination unit 31 determines that the phase current command value Ix when the rotation angle θ of the detected motor 12 is within any current limit range when the energization failure occurs. In order to cope with the fluctuation of the q-axis current deviation ΔIq caused by the restriction of *, the threshold value of the determination process based on the q-axis current deviation ΔIq is changed (see FIGS. 13 to 15).

  More specifically, the abnormality determination unit 31 according to the present embodiment, when the energization failure occurs, if the detected rotation angle θ is within the current limit range, the q axis generated by limiting the phase current command value Ix *. A correction term β corresponding to the fluctuation of the current deviation ΔIq is calculated.

  Specifically, the abnormality determination unit 31 of the present embodiment solves the following equations (23) to (34) according to the detected energization failure occurrence phase and the sign of the q-axis current command value Iq *. Thus, the correction term β for changing the threshold value of the determination process based on the q-axis current deviation ΔIq is calculated.

For the rotation angles θ1, θ2, θ3, and θ4 that define the current limiting range, see FIGS. 13 to 15 and the above equations (5) to (22).
Then, based on a value obtained by adding the correction term β to the normal threshold value I1 (or −I1), a determination process for detecting an abnormality of the control system is executed, thereby enabling more accurate abnormality detection. It is configured to be possible.

Next, a processing procedure for abnormality detection of the control system in the present embodiment will be described.
In the flowchart of FIG. 16 referred to below, the processing of step 501 to step 513 is the same as the processing of step 401 to step 413 in the flowchart of FIG. 12 described above showing the aspect of the first embodiment. For this reason, for the convenience of explanation, the explanation of the processing of Step 501 to Step 513 is omitted.

  As shown in the flowchart of FIG. 16, in the abnormality determination unit 31 of the present embodiment, the rotation angle θ is within one of the current limit ranges calculated in step 512 during two-phase driving (step 505: YES). (Θ1 <θ <θ2, or θ3 <θ <θ4, step 513: NO), first, according to the above equations (23) to (34), the non-energized phase and the rotation angle in the two-phase drive A correction term β corresponding to θ is calculated (step 514).

  Next, the abnormality determination unit 31 determines whether the q-axis current deviation ΔIq is equal to or greater than a value obtained by adding the correction term β to a value (+ I1) in which the sign of the threshold value I1 is positive (step 515). When the current deviation ΔIq is smaller than the corrected value (step 515: NO), it is determined whether the current deviation ΔIq is equal to or smaller than a value obtained by adding the correction term β to a value (−I1) in which the sign of the threshold value I1 is negative. (Step 516). In step 515, the q-axis current deviation ΔIq is equal to or larger than the corrected value (ΔIq ≧ I1 + β, step 515: YES), or in step 516, the q-axis current deviation ΔIq is equal to or smaller than the corrected value (ΔIq If it is determined that ≦ −I1 + β, Step 516: YES, the processing of Steps 507 and 508 is executed. That is, the measurement flag is set to “ON” (step 507), and the counter is incremented (t = t + 1, step 508).

  On the other hand, if it is determined in step 516 that the q-axis current deviation ΔIq is larger than the corrected value (ΔIq> −I1 + β, step 516: NO), that is, if the q-axis current deviation ΔIq is in an appropriate range. When the determination is made, the abnormality determination unit 31 sets the measurement flag to “OFF” (step 517).

  And the abnormality determination part 31 performs the process of step 510, and when the state which shows the said abnormality continues for the predetermined time or more (step 510: YES), it determines with abnormality having generate | occur | produced in the control system (step) 511).

  As described above, according to this embodiment, it is possible to avoid the occurrence of erroneous detection in the current limiting range (θ1 <θ <θ2, θ3 <θ <θ4) in the vicinity of the predetermined rotation angles θA and θB corresponding to the asymptote. . As a result, the control system abnormality can be detected with higher accuracy.

(Third embodiment)
Hereinafter, a third embodiment in which the present invention is embodied in an electric power steering apparatus (EPS) will be described with reference to the drawings. Note that the main difference between the present embodiment and the first embodiment is only the manner of detecting abnormality in the control system. For this reason, for convenience of explanation, the same parts as those in the first embodiment are denoted by the same reference numerals, and the explanation thereof is omitted.

  The abnormality determination unit 31 according to the present embodiment, when energization failure occurs, assumes that the feedback control in the d / q coordinate system is to be executed. A virtual d-axis current command value Id * that can generate a phase current that changes into a secant curve or a cosecant curve with the angles θA and θB as asymptotic lines is calculated. In other words, the abnormality determination unit 31 calculates a virtual d-axis current command value Id * corresponding to the phase current command value Ix * during two phases by the phase current feedback control. Then, a deviation between the d-axis current command value Id * and the actual d-axis current value Id, that is, a d-axis current deviation ΔId is calculated, and the d-axis current deviation ΔId is compared with a predetermined threshold (I2). By doing so, it becomes the structure which detects abnormality of a control system.

  That is, as in the present embodiment, when a phase current that changes into a secant curve or a cosecant curve with a predetermined rotation angle θA, θB as asymptotic lines is passed for each phase other than the energization failure occurrence phase (FIG. 4). The d / q coordinate system includes a d-axis current (Id) that changes in a tangent curve (tangent) with predetermined rotation angles θA and θB as asymptotic lines as shown in FIGS. 17 to 19. appear. Therefore, when two-phase driving is performed, a phase current that changes into a secant curve or a cosecant curve with a predetermined rotation angle θA, θB as asymptotic lines is generated in the two energized phases by executing feedback control in the d / q coordinate system. In order to achieve this, it is only necessary to calculate the d-axis current command value Id * that changes in a tangent curve with predetermined rotation angles θA and θB as asymptotic lines as shown in FIGS.

  That is, even during the two-phase driving, the d-axis current value Id regularly changes according to the rotation angle θ. Therefore, by monitoring this change in the d-axis current value Id, it is possible to determine whether or not an abnormality has occurred in the control system, as in the case of monitoring the change in the q-axis current. In this embodiment, the virtual d-axis current command value Id * is calculated as described above, and is the followability of the actual d-axis current value Id with respect to the d-axis current command value Id *, that is, the deviation thereof. Based on whether or not the d-axis current deviation ΔId is within an appropriate range (± I2), the abnormality detection of the control system is executed.

  Here, in the present embodiment, as described above, when the two-phase driving is performed, the rotation angle θ is in the current limiting range in the vicinity of the predetermined rotation angles θA and θB (θ1 <θ <θ2, θ3 <θ <θ4). The guard process for limiting the phase current command value within a predetermined range is executed. That is, within these current limiting ranges, the d-axis current value Id also does not change in a tangent curve shape with predetermined rotation angles θA and θB as asymptotic lines. Therefore, when the rotation angle θ is within any current limit range, the abnormality determination unit 31 of the present embodiment considers the result of the guard process executed within the current limit range, that is, the phase current command value. Based on the limitation of Ix *, the virtual d-axis current command value Id * is calculated.

  Specifically, the abnormality determination unit 31 of the present embodiment has a rotation angle θ other than a current limit range in the vicinity of predetermined rotation angles θA and θB corresponding to asymptotic lines (0 <θ ≦ θ1, θ2 ≦ θ ≦ θ3). (θ4 ≦ θ ≦ 2π), the following formulas (35) to (37) are solved in accordance with the energization failure occurrence phase to form a tangent curve with the predetermined rotation angles θA and θB as asymptotic lines. A changing virtual d-axis current command value Id * is calculated.

For the rotation angles θ1, θ2, θ3, and θ4 that define the current limiting range, see FIGS. 13 to 15 and the above equations (5) to (22).
When the rotation angle θ is in any current limiting range (θ1 <θ <θ2, θ3 <θ <θ4), depending on the energization failure occurrence phase at that time and the sign of the q-axis current command value Iq * The virtual d-axis current command value Id * is calculated by solving the following equations (38) to (49).

  Then, the deviation between the virtual d-axis current command value Id * obtained by the above equations (35) to (49) and the actual d-axis current value Id, that is, the d-axis current deviation ΔId and a predetermined threshold value I2 Based on the comparison, a determination process for detecting an abnormality of the control system is executed, so that the abnormality can be detected with higher accuracy.

Next, the control system abnormality detection mode in this embodiment will be described.
In the flowchart of FIG. 20 referred to below, the processing of step 601 to step 612 is the same as the processing of step 401 to step 412 in the above-described flowchart of FIG. 12 showing the aspect of the first embodiment (step Only the state quantities (Id *, Id) acquired in 601 are different). For this reason, for the convenience of explanation, the explanation of the processing of Step 601 to Step 612 is omitted.

  As shown in the flowchart of FIG. 20, the abnormality determination unit 31 of the present embodiment calculates the current limit range in step 512 during two-phase driving (step 605: YES), and then the calculation result and energization at that time. Based on the failure occurrence phase and the sign of the q-axis current command value Iq *, a virtual d-axis current command value Id * is calculated from the above equations (35) to (49) (step 613). Next, the abnormality determination unit 31 calculates a deviation between the virtual d-axis current command value Id * calculated in step 613 and the actual d-axis current value Id, that is, a d-axis current deviation ΔId (step 614). ), It is determined whether or not the calculated d-axis current deviation ΔId (absolute value thereof) is equal to or greater than a predetermined threshold value I2 (step 615). If it is determined in step 615 that the d-axis current deviation ΔId (absolute value) is equal to or greater than a predetermined threshold value I2 (ΔId ≧ I2, step 615: YES), the processing in steps 607 and 608 is executed. To do. That is, the measurement flag is set to “ON” (step 607), and the counter is incremented (t = t + 1, step 608).

  On the other hand, when it is determined in step 615 that the d-axis current deviation ΔId (absolute value thereof) is less than the predetermined threshold I2 (ΔId <I2, step 615: NO), that is, the d-axis current deviation ΔId is in an appropriate range. If it is determined that the error is present, the abnormality determination unit 31 sets the measurement flag to “OFF” (step 616).

  And the abnormality determination part 31 performs the process of step 610, and when the state which shows the said abnormality continues more than predetermined time (step 610: YES), it determines with the abnormality having generate | occur | produced in the control system (step) 611).

  As described above, according to this embodiment, it is possible to avoid the occurrence of erroneous detection in the current limiting range (θ1 <θ <θ2, θ3 <θ <θ4) in the vicinity of the predetermined rotation angles θA and θB corresponding to the asymptote. . As a result, the control system abnormality can be detected with higher accuracy.

(Fourth embodiment)
Hereinafter, a fourth embodiment in which the present invention is embodied in an electric power steering apparatus (EPS) will be described with reference to the drawings. The main difference between the present embodiment and the third embodiment is only the manner of detecting an abnormality in the control system. For this reason, for convenience of explanation, the same parts as those of the third embodiment are denoted by the same reference numerals, and the description thereof is omitted.

  As in the third embodiment, the abnormality determination unit 31 according to the present embodiment calculates a virtual d-axis current command value Id * and then further calculates the virtual d-axis current command value Id * and q. A composite vector Idq * between the shaft current command value Iq * is calculated (see FIGS. 21 to 23). Note that a composite vector Idq * of the virtual current command value in the d / q coordinate system and a composite vector of real current values in the d / q coordinate system described later (combination of the d-axis current value Id and the q-axis current value Iq). Needless to say, the vector) Idq is the square root of the sum of squares of the d-axis current command value Id * and the q-axis current command value Iq *, and the d-axis current value Id and the q-axis current value Iq, respectively. Then, the abnormality detection of the control system is executed by monitoring the change of the composite vector Idq *.

  More specifically, the abnormality determination unit 31 according to the present embodiment determines whether or not the rotation angle θ is in any current limiting range (θ1 <θ <θ2, θ3 <θ <θ4) and the current-carrying failure occurrence phase at that time. Accordingly, the following equations (50) to (53) are solved to calculate a combined vector Idq * of the virtual current command value in the d / q coordinate system.

For the rotation angles θ1, θ2, θ3, and θ4 that define the current limiting range, see FIGS. 13 to 15 and the above equations (5) to (22).
Then, the resultant vector Idq * of the virtual current command value in the d / q coordinate system and the resultant vector of the actual current value in the d / q coordinate system (d-axis current value Id and q-axis current value Iq). The deviation ΔIdq from the combined vector) Idq is calculated, and the deviation ΔIdq is compared with a predetermined threshold value (I3) to detect abnormality of the control system.

Next, the control system abnormality detection mode in this embodiment will be described.
In the flowchart of FIG. 24 to be referred to below, the processing of step 701 to step 713 is the same as the processing of step 601 to step 613 in the flowchart of FIG. 20 described above showing the aspect of the third embodiment. For this reason, for the convenience of explanation, the explanation of the processing of Step 701 to Step 713 is omitted.

  As shown in the flowchart of FIG. 24, the abnormality determination unit 31 of the present embodiment calculates a virtual d-axis current command value Id * in step 713 as in the third embodiment. A combined vector Idq * of the axis current command value Id * and the q-axis current command value Iq * is calculated (step 714).

  Next, the abnormality determination unit 31 combines a virtual current command value composite vector Idq * in the d / q coordinate system and a real current value composite vector (d-axis current value Id and q-axis in the d / q coordinate system). A deviation ΔIdq from the combined vector (Idq) of the current value Iq is calculated (ΔIdq = Idq * −Idq, step 715), and it is determined whether or not the deviation ΔIdq is equal to or greater than a predetermined threshold I3 (step 716). If it is determined in step 716 that the deviation ΔIdq (the absolute value thereof) is equal to or greater than a predetermined threshold value I3 (ΔIdq ≧ I3, step 716: YES), the processing in steps 707 and 708 is executed. That is, the measurement flag is set to “ON” (step 707), and the counter is incremented (t = t + 1, step 708).

  On the other hand, when it is determined in step 716 that the combined vector deviation ΔIdq (absolute value thereof) is less than the predetermined threshold value I3 (ΔIdq <I3, step 716: NO), the combined vector deviation ΔIdq is in an appropriate range. If it is determined that the error is present, the abnormality determination unit 31 sets the measurement flag to “OFF” (step 717).

  And the abnormality determination part 31 performs the process of step 710, and when the state which shows the said abnormality continues more than predetermined time (step 710: YES), it determines with the abnormality having generate | occur | produced in the control system (step) 711).

As described above, according to the present embodiment, it is possible to obtain the same operations and effects as those of the third embodiment.
(Fifth embodiment)
Hereinafter, a fifth embodiment in which the present invention is embodied in an electric power steering apparatus (EPS) will be described with reference to the drawings. Note that the main difference between the present embodiment and the first embodiment is only the manner of detecting abnormality in the control system. For this reason, for convenience of explanation, the same parts as those in the first embodiment are denoted by the same reference numerals, and the explanation thereof is omitted.

  The abnormality determination unit 31 of the present embodiment selects one of the two energized phases as a determination phase during two-phase driving. Then, the abnormality determination of the control system is executed by monitoring the phase current deviation in the determination phase.

  That is, by executing the phase current feedback control based on the phase current command value that changes into a secant curve or a cosecant curve with the predetermined rotation angles θA and θB corresponding to the energization failure occurrence phase as asymptotic lines, The phase current value Iy continuously changes following the phase current command value Iy * (except for predetermined rotation angles θA and θB) unless an abnormality has occurred in the control system. Therefore, by comparing these deviations, that is, the phase current deviation ΔIy and the predetermined threshold value I3, it is possible to detect an abnormality of the control system with high accuracy even during two-phase driving. This also applies to the current limiting range (θ1 <θ <θ2, θ3 <θ <θ4) in the vicinity of the predetermined rotation angles θA and θB corresponding to the asymptote, and this embodiment uses these characteristics. By doing so, it is possible to detect abnormality with high accuracy over the entire rotation region.

Next, the control system abnormality detection mode in this embodiment will be described.
In the flowchart of FIG. 25 to be referred to below, the processing of step 801 to step 811 is the same as the processing of step 401 to step 411 in the flowchart of FIG. 12 described above showing the aspect of the first embodiment (step Only the state quantities (Iu, Iv, Iw, Iu *, Iv *, Iw *) acquired in 801 are different). For this reason, for the convenience of description, the description of the processing of Step 801 to Step 811 is omitted.

  As shown in the flowchart of FIG. 25, when the abnormality determination unit 31 determines in step 805 that the two-phase driving is already in progress (step 805: YES), one of the two-phase energized phases is set as the determination phase. Next, the phase current deviation ΔIy in the determination phase is calculated (ΔIy = Iy * −Iy, step 813). In the present embodiment, of the two energized phases, the phase that is the determination phase is set in advance for each energized failure phase. Then, it is determined whether or not the phase current deviation ΔIy (absolute value) is greater than or equal to a predetermined threshold I4 (step 814), and the phase current deviation ΔIy (absolute value) is greater than or equal to a predetermined threshold I4. For (| ΔIy | ≧ I4, Step 814: YES), the processes of Steps 807 and 808 are executed. That is, the measurement flag is set to “ON” (step 807), and the counter is incremented (t = t + 1, step 808).

  On the other hand, in step 814, when the phase current deviation ΔIy (absolute value thereof) is less than the predetermined threshold I4 (| ΔIy | <I4, step 814: NO), that is, when it is determined that it is in the appropriate range, The abnormality determination unit 31 sets the measurement flag to “OFF” (step 815).

  And the abnormality determination part 31 performs the process of step 810, and when the state which shows the said abnormality continues for the predetermined time or more (step 810: YES), it determines with the abnormality having generate | occur | produced in the control system (step) 811).

  As described above, according to this embodiment, it is possible to avoid the occurrence of erroneous detection in the current limiting range (θ1 <θ <θ2, θ3 <θ <θ4) in the vicinity of the predetermined rotation angles θA and θB corresponding to the asymptote. . As a result, the control system abnormality can be detected with higher accuracy.

(Sixth embodiment)
Hereinafter, a fifth embodiment in which the present invention is embodied in an electric power steering apparatus (EPS) will be described with reference to the drawings. Note that the main difference between the present embodiment and the first embodiment is only the manner of detecting abnormality in the control system. For this reason, for convenience of explanation, the same parts as those in the first embodiment are denoted by the same reference numerals, and the explanation thereof is omitted.

  The abnormality determination unit 31 of the present embodiment monitors the current values (Iy1, Iy2) of the respective phases that are energized phases during two-phase driving. Then, based on the comparison between the total value Iz (absolute value thereof) and a predetermined threshold value I5, the abnormality detection of the control system is executed.

  That is, according to Kirchhoff's law, as long as no abnormality occurs in the control system, the total value Iz of the two-phase phase current values Iy1 and Iy2 to be energized phases should be “0”. Therefore, when the total value Iz (absolute value thereof) exceeds the appropriate range and deviates from “0”, it can be determined that an abnormality has occurred in the control system. And this embodiment becomes a structure which can detect abnormality with high precision over the whole rotation area | region by utilizing such a characteristic.

  Specifically, as shown in the flowchart of FIG. 26, when the abnormality determination unit 31 determines in step 905 that the two-phase driving has already been performed (step 905: YES), the two-phase phase current that becomes the energized phase is determined. The total value Iz of the values Iy1 and Iy2 is calculated (Iz = Iy1 + Iy2, step 912), and it is determined whether or not the total value Iz (absolute value thereof) is equal to or greater than a predetermined threshold value I5 (step 913). If the total value ΔIz (absolute value thereof) is equal to or greater than a predetermined threshold value I5 (| Iz | ≧ I5, step 913: YES), the processing of steps 907 and 908 is executed. That is, the measurement flag is set to “ON” (step 907), and the counter is incremented (t = t + 1, step 908).

  On the other hand, when the total value Iz (the absolute value thereof) of the two-phase phase current values Iy1 and Iy2 serving as the energized phase is less than the predetermined threshold value I5 in the step 913 (| Iz | <I5, step 913: NO) In other words, when it is determined that it is within the appropriate range, the abnormality determination unit 31 sets the measurement flag to “OFF” (step 914).

  And the abnormality determination part 31 performs the process of step 910, and when the state which shows the said abnormality continues more than predetermined time (step 910: YES), it determines with abnormality having generate | occur | produced in the control system (step) 911).

  Note that the processing from step 901 to step 911 is the same as the processing from step 401 to step 411 in the flowchart of FIG. 12 showing the aspect of the first embodiment (state quantities (Iu, Iv acquired in step 901) Only Iw) is different). For this reason, for the convenience of description, the description of the processing of Step 901 to Step 911 is omitted.

  As described above, according to the present embodiment, it is possible to accurately detect an abnormality in the control system even during two-phase driving. In particular, even in a current limiting range (θ1 <θ <θ2, θ3 <θ <θ4) in the vicinity of the predetermined rotation angles θA and θB corresponding to the asymptote, the occurrence of erroneous detection due to the setting of the current limiting range is avoided. This makes it possible to detect anomalies with high accuracy over the entire rotation region. In addition, such a characteristic has an advantage that it is effective even when two-phase driving is performed by a method other than the phase current feedback as in the present embodiment.

In addition, you may change each said embodiment as follows.
In each of the above embodiments, the present invention is embodied in an electric power steering device (EPS), but may be embodied in a motor control device used for applications other than EPS.

  In the present embodiment, the ECU 11 as the motor control device is roughly divided into three control modes: “normal control mode”, “assist stop mode”, and “two-phase drive mode”. However, the mode of motor control when an abnormality occurs is not limited to these modes. In other words, any configuration may be applied as long as the motor control is executed by using two phases other than the energization failure phase as the energization phase when the energization failure phase occurs. Further, the method of abnormality detection (determination) is not limited to the configuration of each of the above embodiments.

  In each of the above embodiments, the current command value calculation unit 23 outputs a phase current command value for one of the two phases other than the energization failure occurrence phase during two-phase driving, and the motor control signal generation unit 24 After calculating the phase voltage command value for the phase, the phase voltage command value for the other phase is calculated based on the calculated phase voltage command value. However, the present invention is not limited to this, and the current command value calculation unit 23 may output phase current command values for both two phases other than the energization failure occurrence phase.

  In each of the above embodiments, based on the above equations (1) to (3), when the U phase or W phase is abnormal, the V phase phase current command value Iv * is calculated, and when the V phase is abnormal, The U-phase phase current command value Iu * is calculated. However, not limited to this, the W-phase current command value (Iw *) is calculated when the U-phase or V-phase is abnormal, and the V-phase current command value (Iv *) is calculated when the W-phase is abnormal. It is good also as composition to do. In addition, each phase current command value in this case can be calculated by reversing the signs of the above formulas (1) to (3).

  Furthermore, the phase current command value at the time of occurrence of an abnormality does not necessarily have to be completely the same as that calculated by the above equations (1) to (3). That is, even if a phase current command value that changes to a substantially secant curve or a substantially cosecant curve with an asymptotic line as a predetermined rotation angle, or changes in an approximate manner, is similar to each of the above embodiments. An effect can be obtained. However, when the phase current command value is calculated based on the above equations (1) to (3), it is possible to generate a motor current closest to the required torque, and the phase current command calculated based on each equation Needless to say, a method that calculates a value close to a value provides a more remarkable effect.

  In each of the above embodiments, the control system abnormality determination is normally performed on the basis of the q-axis current deviation ΔIq. However, the present invention is not limited to this, and the invention shown in each of the above embodiments is applied to the case where the abnormality determination of the control system in the normal time is performed based on the d-axis current deviation ΔId or the combined vector deviation in the d / q coordinate system. May be.

  In particular, in the invention of the second embodiment described above, in the case where the d-axis current deviation ΔId is based on the normal state, the threshold value related to the d-axis current deviation ΔId is set based on the combined vector deviation. The threshold for the combined vector deviation may be changed.

  Further, in the third embodiment, the d-axis current command is used as a virtual current command value in the d / q coordinate system corresponding to the phase current command value Ix * used for determining the control system abnormality during two-phase driving. In the fourth embodiment, the value Id * is calculated as a combined vector Idq *, and the d-axis current deviation ΔId or the combined vector deviation ΔIdq is used as a basis for determining the abnormality of the control system. However, the present invention is not limited to this, and a q-axis current command is obtained as a virtual current command value in the d / q coordinate system corresponding to the phase current command value Ix *, and the q-axis current deviation is used as a basis for abnormality determination of the control system. It is good also as a structure.

  In the fourth embodiment, the above formulas (50) to (53) for calculating the combined vector Idq * of the virtual current command value in the d / q coordinate system arrange the square root of the square sum. However, there is no “d-axis current command value Id *” in these equations. Therefore, when using these equations, it goes without saying that it is not always necessary to calculate the d-axis current command value Id * in advance.

The schematic block diagram of an electric power steering device (EPS). The block diagram which shows the electric constitution of EPS. The flowchart which shows the process sequence of an electricity supply failure phase detection. Explanatory drawing which shows transition of each phase electric current at the time of two-phase drive (at the time of U-phase electricity failure). Explanatory drawing which shows transition of each phase current at the time of two-phase drive (at the time of V-phase electricity supply failure). Explanatory drawing which shows transition of each phase current at the time of two-phase drive (at the time of W-phase electricity supply failure). Explanatory drawing which shows transition of the q-axis current (deviation) at the time of two-phase drive (at the time of U-phase electricity failure). Explanatory drawing which shows transition of the q-axis current (deviation) at the time of two-phase drive (at the time of V-phase energization failure). Explanatory drawing which shows transition of q-axis current (deviation) at the time of two-phase drive (at the time of W-phase energization failure). The flowchart which shows the process sequence of abnormality determination and control mode switching. The flowchart which shows the process sequence of the motor control signal generation at the time of two-phase drive. The flowchart which shows the process sequence about abnormality determination of the control system in 1st Embodiment. Explanatory drawing which shows transition of the q-axis current deviation at the time of the two-phase drive (at the time of U-phase electricity failure) in 2nd Embodiment. Explanatory drawing which shows transition of the q-axis current deviation at the time of the two-phase drive (at the time of V-phase electricity supply failure) in 2nd Embodiment. Explanatory drawing which shows transition of the q-axis current deviation at the time of the two-phase drive (at the time of W-phase electricity supply failure) in 2nd Embodiment. The flowchart which shows the process sequence about abnormality determination of the control system in 2nd Embodiment. Explanatory drawing which shows transition of d-axis current at the time of two-phase drive (at the time of U-phase conduction failure). Explanatory drawing which shows transition of the d-axis current at the time of two-phase drive (at the time of V-phase conduction failure). Explanatory drawing which shows transition of d-axis current at the time of two-phase drive (at the time of W-phase energization failure). The flowchart which shows the process sequence about abnormality determination of the control system in 3rd Embodiment. Explanatory drawing which shows transition of the d / q-axis synthetic vector at the time of two-phase drive (at the time of U-phase conduction failure). Explanatory drawing which shows transition of the d / q-axis synthetic vector at the time of two-phase drive (at the time of V-phase energization failure). Explanatory drawing which shows transition of the d / q-axis synthetic | combination vector at the time of two-phase drive (at the time of W-phase electricity supply failure). The flowchart which shows the process sequence about abnormality determination of the control system in 4th Embodiment. The flowchart which shows the process sequence about abnormality determination of the control system in 5th Embodiment. The flowchart which shows the process sequence about abnormality determination of the control system in 6th Embodiment. Explanatory drawing which shows the aspect of the two-phase drive which uses the two phases other than the conventional energization failure generation phase as an energization phase. Explanatory drawing which shows transition of the d-axis current and q-axis current at the time of the conventional two-phase drive.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Electric power steering apparatus (EPS), 10 ... EPS actuator, 11 ... ECU, 12 ... Motor, 12u, 12v, 12w ... Motor coil, 17 ... Microcomputer, 18 ... Drive circuit, 23 ... Current command value calculating part, 24 ... motor control signal generation unit, 24a ... first control unit, 24b ... second control unit, 29 ... switching control unit, 31 ... abnormality determination unit, Ix, Iy, Iy1, Iy2, Iu, Iv, Iw ... phase current value , Iz ... total value, Ix *, Iu *, Iv *, Iw * ... phase current command value, Ix_max ... maximum value, Vx *, Vu *, Vv *, Vw * ... phase voltage command value, Id ... d-axis current Value, Iq ... q-axis current value, Id * ... d-axis current command value, Iq * ... q-axis current command value, ΔId ... d-axis current deviation, ΔIq ... q-axis current deviation, Idq, Idq * ... composite vector, ΔIdq ... Deviation, I1, I2, I3, I4, I5 ... q-axis current command value, β ... correction term, θ, θA, θB, θ1, θ2, θ3, θ4... rotation angle.

Claims (9)

A motor control signal output means for outputting a motor control signal; and a drive circuit for supplying three-phase drive power to the motor based on the motor control signal. The motor control signal output means calculates a current command value. Current command value calculating means for generating, motor control signal generating means for generating the motor control signal by executing current feedback control in the d / q coordinate system based on the current command value, and current deviation in the d / q coordinate system. An abnormality detecting means for detecting an abnormality of the control system on the basis of the control system and detecting an energization failure occurring in each phase of the motor, and when the occurrence of the energization failure is detected, other than the energization failure occurrence phase In the motor control device for executing the output of the motor control signal with the two phases as energized phases,
When the occurrence of the energization failure is detected, the current command value calculating means changes the phase current that changes into a secant curve or a remainder curve with a predetermined rotation angle corresponding to the energization failure occurrence phase as an asymptotic line A command value is calculated, and the motor control signal generating means generates the motor control signal by executing phase current feedback control based on the calculated phase current command value,
The motor control signal output means is provided with limiting means for limiting the phase current command value within a predetermined range,
The abnormality detection means does not execute a determination for detecting an abnormality of the control system when the energization failure occurs and the phase current command value is within a rotation angle range that limits the value within a predetermined range. The motor control apparatus characterized by these. A motor control signal output means for outputting a motor control signal; and a drive circuit for supplying three-phase drive power to the motor based on the motor control signal. The motor control signal output means calculates a current command value. Current command value calculating means, motor control signal generating means for generating the motor control signal by executing current feedback control in the d / q coordinate system based on the current command value, current deviation in the d / q coordinate system, An abnormality detecting means capable of detecting an abnormality in the control system by comparison with a predetermined threshold and detecting an energization failure occurring in each phase of the motor, and when the occurrence of the energization failure is detected, In the motor control device that executes the output of the motor control signal with two phases other than the energization failure occurrence phase as the energization phase,
When the occurrence of the energization failure is detected, the current command value calculating means changes the phase current that changes into a secant curve or a remainder curve with a predetermined rotation angle corresponding to the energization failure occurrence phase as an asymptotic line A command value is calculated, and the motor control signal generating means generates the motor control signal by executing phase current feedback control based on the calculated phase current command value,
The motor control signal output means is provided with limiting means for limiting the phase current command value within a predetermined range,
When the energization failure occurs, if the phase current command value is within the rotation angle range that limits the phase current command value within a predetermined range, the fluctuation of the current deviation in the d / q coordinate system caused by the limitation of the phase current command value may occur. A motor control device, characterized in that the threshold value is changed in order to respond. The motor control device according to claim 2,
The current command value calculation means, according to the energization failure occurrence phase, the following equations,

(However, θ: rotation angle, Iq *: q-axis current command value, Iu *: U-phase current command value, Iv *: V-phase current command value)
And calculating the phase current command value based on
The limiting means is:

(However, Ix *: Phase current command value, Ix_max: Maximum value of phase current that can be energized)
Based on the phase current command value within a predetermined range,
The abnormality detection means detects an abnormality of the control system by comparing a q-axis current deviation of the d / q coordinate system with a predetermined threshold value,
When the energization failure occurs, if the phase current command value is within a rotation angle range that limits the phase current command value within a predetermined range,

(Where, β: correction term, Iq *: q-axis current command value, Ix_max: maximum current that can be energized)
Calculating a correction term for changing the threshold by
A motor control device. A motor control signal output means for outputting a motor control signal; and a drive circuit for supplying three-phase drive power to the motor based on the motor control signal. The motor control signal output means calculates a current command value. Current command value calculating means for generating, motor control signal generating means for generating the motor control signal by executing current feedback control in the d / q coordinate system based on the current command value, and current deviation in the d / q coordinate system. An abnormality detecting means for detecting an abnormality of the control system on the basis of the control system and detecting an energization failure occurring in each phase of the motor, and when the occurrence of the energization failure is detected, other than the energization failure occurrence phase In the motor control device for executing the output of the motor control signal with the two phases as energized phases,
When the occurrence of the energization failure is detected, the current command value calculating means changes the phase current that changes into a secant curve or a remainder curve with a predetermined rotation angle corresponding to the energization failure occurrence phase as an asymptotic line A command value is calculated, and the motor control signal generating means generates the motor control signal by executing phase current feedback control based on the calculated phase current command value,
The motor control signal output means is provided with limiting means for limiting the phase current command value within a predetermined range,
The abnormality detecting means calculates a virtual current command value of the d / q coordinate system corresponding to the phase current command value in consideration of the limitation when the energization failure occurs, and the virtual current command value And detecting an abnormality of the control system based on a deviation between the actual current value of the d / q coordinate system,
A motor control device. The motor control device according to claim 4,
The current command value calculation means, according to the energization failure occurrence phase, the following equations,

(However, θ: rotation angle, Iq *: q-axis current command value, Iu *: U-phase current command value, Iv *: V-phase current command value)
The phase current command value is calculated based on
The abnormality detection means includes the following expressions:

(However, Id *: d-axis current command value, Iq *: q-axis current command value)
And calculating the d-axis current command value as the virtual current command value, and the limiting means is:

(However, Ix *: Phase current command value, Ix_max: Maximum value of phase current that can be energized)
The phase current command value is limited within a predetermined range based on
When the abnormality detection means is within the rotation angle range that limits the phase current command value within a predetermined range when the energization failure occurs,

(However, Id *: d-axis current command value, Iq *: q-axis current command value, Ix_max: maximum current that can be energized)
The d-axis current command value is calculated by the motor control device. The motor control device according to claim 4,
The current command value calculation means, according to the energization failure occurrence phase, the following equations,

(However, θ: rotation angle, Iq *: q-axis current command value, Iu *: U-phase current command value, Iv *: V-phase current command value)
The phase current command value is calculated based on
The abnormality detection means includes the following expressions:

(However, Idq *: vector of current command values in the d / q coordinate system, Iq *: q-axis current command value)
To calculate a combined vector of current command values in the d / q coordinate system as the virtual current command value,
The limiting means is:

(However, Ix *: Phase current command value, Ix_max: Maximum value of phase current that can be energized)
The phase current command value is limited within a predetermined range based on
When the abnormality detection means is within the rotation angle range that limits the phase current command value within a predetermined range when the energization failure occurs,

(However, Idq *: Composite vector of current command values in the d / q coordinate system, Ix_max: Maximum current that can be energized)
Calculating a combined vector of current command values in the d / q coordinate system by
A motor control device. A motor control signal output means for outputting a motor control signal; and a drive circuit for supplying three-phase drive power to the motor based on the motor control signal. The motor control signal output means calculates a current command value. Current command value calculating means, motor control signal generating means for generating the motor control signal by executing current feedback control in the d / q coordinate system based on the current command value, control system abnormality and each phase of the motor And an abnormality detection means capable of detecting a failure in energization, and when the occurrence of the energization failure is detected, the motor control signal is output with two phases other than the failure phase as the energization phase. In the motor control device
When the occurrence of the energization failure is detected, the current command value calculating means changes the phase current that changes into a secant curve or a remainder curve with a predetermined rotation angle corresponding to the energization failure occurrence phase as an asymptotic line A command value is calculated, and the motor control signal generating means generates the motor control signal by executing phase current feedback control based on the calculated phase current command value,
The abnormality detecting means, when the energization failure occurs, executing an abnormality determination of the control system based on a phase current deviation for at least one of the energized phases;
A motor control device. A motor control signal output means for outputting a motor control signal; and a drive circuit for supplying three-phase drive power to the motor based on the motor control signal. The motor control signal output means calculates a current command value. Current command value calculation means for performing, motor control signal generation means for generating the motor control signal by execution of current feedback control based on the current command value, abnormalities in the control system and energization failures occurring in each phase of the motor In a motor control device that includes a detectable abnormality detection means, and when the occurrence of the energization failure is detected, executes the output of the motor control signal with two phases other than the energization failure occurrence phase as the energization phase,
The abnormality detecting means, when the energization failure occurs, executing an abnormality determination of the control system based on a total value of both-phase current values constituting the energized phase;
A motor control device.   An electric power steering device comprising the motor control device according to any one of claims 1 to 8.

Images